Re: [PATCH] usb: typec: fusb302: Fix an undefined reference to 'extcon_get_state'

[Guenter Roeck <linux@roeck-us.net>]

On Thu, Dec 12, 2019 at 02:49:37PM +0200, Heikki Krogerus wrote:
> On Thu, Dec 12, 2019 at 01:20:53PM +0100, Greg KH wrote:
> > On Thu, Dec 12, 2019 at 01:18:05PM +0200, Heikki Krogerus wrote:
> > > On Thu, Dec 12, 2019 at 05:40:40PM +0800, zhong jiang wrote:
> > > > On 2019/12/12 17:28, Greg KH wrote:
> > > > > On Thu, Dec 12, 2019 at 05:15:28PM +0800, zhong jiang wrote:
> > > > >> On 2019/12/12 17:01, Heikki Krogerus wrote:
> > > > >>> On Thu, Dec 12, 2019 at 03:34:23PM +0800, zhong jiang wrote:
> > > > >>>> Fixes the following compile error:
> > > > >>>>
> > > > >>>> drivers/usb/typec/tcpm/fusb302.o: In function `tcpm_get_current_limit':
> > > > >>>> fusb302.c:(.text+0x3ee): undefined reference to `extcon_get_state'
> > > > >>>> fusb302.c:(.text+0x422): undefined reference to `extcon_get_state'
> > > > >>>> fusb302.c:(.text+0x450): undefined reference to `extcon_get_state'
> > > > >>>> fusb302.c:(.text+0x48c): undefined reference to `extcon_get_state'
> > > > >>>> drivers/usb/typec/tcpm/fusb302.o: In function `fusb302_probe':
> > > > >>>> fusb302.c:(.text+0x980): undefined reference to `extcon_get_extcon_dev'
> > > > >>>> make: *** [vmlinux] Error 1
> > > > >>> There are stubs for those functions so that really should not be
> > > > >>> happening. I can not reproduce that.
> > > > >> It can be reproduced in next branch. you can try it in the latest next branch.
> > > > > Can it be reproduced in 5.5-rc1?
> > > > >
> > > > commit 78adcacd4edbd6795e164bbda9a4b2b7e51666a7
> > > > Author: Stephen Rothwell <sfr@canb.auug.org.au>
> > > > Date:   Thu Dec 12 15:48:07 2019 +1100
> > > > 
> > > >     Add linux-next specific files for 20191212
> > > > 
> > > > I  reproduce it  based on this commit.  The related config is attached.
> > > 
> > > OK, now I get what's going on. EXTCON is build as a module, but
> > > FUSB302 is not. This should be explained in the commit message.
> > > 
> > > That does not mean we have to force everybody to enable EXTCON in
> > > order to use this driver. Try something like this:
> > > 
> > > diff --git a/drivers/usb/typec/tcpm/Kconfig b/drivers/usb/typec/tcpm/Kconfig
> > > index 72481bbb2af3..06e026f6325c 100644
> > > --- a/drivers/usb/typec/tcpm/Kconfig
> > > +++ b/drivers/usb/typec/tcpm/Kconfig
> > > @@ -31,6 +31,7 @@ endif # TYPEC_TCPCI
> > > 
> > >  config TYPEC_FUSB302
> > >         tristate "Fairchild FUSB302 Type-C chip driver"
> > > +       depends on EXTCON=n || EXTCON=y || (EXTCON=m && m)
> > 
> > Ugh.  We need a better "pattern" for stuff like this, it's getting more
> > and more frequent.
> > 
> > And no, I don't have a better idea :(
> 
> I think this works:
> 
>         depends on EXTCON || !EXTCON

Yes, that is what I usually use for hwmon, though it dos look weird.
I agree, it would be nice to have something expressing such dependencies
in a better way.

Guenter